This grant proposes to identify medical and other individual characteristics of persons aged over 64 years that put them at increased risk of dying due to weather, to identify interactions with air pollution that contribute to that risk, and to identify profiles of patterns of pollutants and weather parameters that are particularly associated with elevated risk. Further, we will identify characteristics of community, such as socio-economic status, percent of impermeable surface, of green space, of water, climate zone, variability of weather, air conditioning prevalence, behavioral risk, baseline disease rates, etc which modify the risk of dying, and finally, interactions between the community level and individual characteristics. Importantly, these community level characteristics will be defined on the zip code level, not the city level, allowing us to capture the impact of true local land use. In addition, examining a less explored weather parameter, we will examine the association of rainfall with hospital admissions for gastrointestinal illness in the elderly. Finally, we will conduct a pilot risk assessment using projections, again on a fine scale, of the distribution of weather and pollution in 2030 compared to today, as well as community level projects of changes in risk modifiers. The results of this analysis will aid NIH by identifying disease states that convey risk, and more broadly aid the task of identifying interventions that can improve public health by reducing risk, and where, geographically, those interventions will be most efficacious. The methods used will be case-crossover and case-only analyses to identify risk and risk modifiers, and k-clustering to group weather and air pollution parameters.